This page is the serious scientific surface of MoscaQuant. It separates experimental claims from presentation, preserves uncertainty, and links directly to public code, test playbacks, provenance, and review channels.
MoscaQuant investigates whether a connectome-inspired autonomous system can produce structured, measurable behavior in a constrained market environment without engineering a preferred outcome into the subject.
The public science surface focuses on contracts, telemetry, provenance, replay, motor/physics boundaries, and reproducible evidence. The production visualization layer is deliberately downstream of those records.
Pair-specific downstream convergence exceeded the null reference in the completed run. This is an observation from the MORTY/LILITH pair, not a population-level claim.
Pair-specific HHI exceeded the null reference in the completed run. Interpretation remains bounded to this implementation and pair.
MQ-3 zero-crossing and the SQ-04 propagation ceiling remain unresolved. They are preserved as open questions rather than silently tuned away.
Neuroscope provides playback-oriented views for inspecting recorded test behavior and experimental state without relying on the production Panopticon presentation layer.
Frozen recorded evidence only. Selecting a playback does not run a new experiment.
No experiments match. Clear the filters or try another name or node ID.
Replays the frozen 192-frame connectome telemetry used by the viewer. Real soma coordinates and topology fallback remain explicitly distinguished.
Watch a recorded sequence through the network. This is playback of existing data, not a brain simulation running in your browser.
Provides the baseline reference for the viewer; playback itself adds no experimental finding.
Inspect the packaged baseline replay across its recorded frames.
Attenuates selected causal-source signals for 13 directed edges at retained dose levels, with deterministic repeats, sham checks and available timing/matched controls.
Turn down one recorded connection at a time and inspect the downstream traces. Repeating a test exactly checks reproducibility; it does not by itself prove a biological explanation.
The artifact reports all 13 tests completed, exact deterministic replicates and shams, and verified dose telemetry. Interpretation remains NOT_YET_INTERPRETED.
Inspect retained 25%, 50%, 75% and 100% attenuation conditions and the associated controls.
Inspects upstream and intermediate attenuation along the retained 56393 → 68045 → 1273 pathway.
Turn down either step in a two-step route and compare the recorded downstream response.
The pathway artifact is classified confirmatory, with interpretation status NOT_YET_INTERPRETED.
Select upstream or intermediate attenuation in the existing viewer and inspect the retained dose conditions.
Compares A-only, B-only and combined attenuation for three systems where two inputs converge on a target.
Turn down either input, or both together, and compare the recorded traces at the shared destination.
The artifact reports all systems completed, exact combined replicates and shams, and verified intervention telemetry. Statistical interaction remains unassigned.
Inspect A-only, B-only and A+B conditions at the retained attenuation levels.
Pair-specific downstream convergence against its null reference.
In this recorded pair, downstream convergence was higher than its comparison reference. That is a result about this pair, not a result about every fly.
Observed 0.5621987198; null median 0.4052166225; p=9.999e-05; ratio 1.3874.
Completed MORTY/LILITH pair-specific comparison, as reported on the existing Science surface.
Pair-specific HHI against its null reference.
This pair had a more concentrated measured pattern than its comparison reference. That does not show the same pattern holds across a population.
Observed 0.0002087914; null median 0.0001640372; p=9.999e-05; ratio 1.2728.
Completed MORTY/LILITH pair-specific comparison, as reported on the existing Science surface.
The evidence path is designed to distinguish requested motor intent from achieved physical state, preserve source ordering, detect replay divergence, and report gaps or backpressure rather than silently repairing the record.
The completed SQ-03 observations are based on n=2 and do not support population claims.
Current results remain tied to the present model, protocol, configuration, and implementation.
Terms such as mood, stress, reward, or courtship are model-state indicators, not evidence of human-like subjective experience.
Panopticon can interpret public state for display, but visual flair cannot create or alter scientific observations.
External review is welcome. Useful feedback includes protocol criticism, statistical concerns, alternative explanations, implementation questions, reproducibility problems, and proposed falsification tests.
Please distinguish scientific criticism from product or visual feedback so each can be tracked in the appropriate repository.